The goal of the proposed work is to develop platform targeting agents for imaging and, eventually, therapy of metastatic melanoma. These agents are heteromultimeric ligands directed against combinations of cell surface receptors or epitopes. Prior work by our team and others has shown that homomultimers of ligands bind with significantly higher affinity compared to their component monomers through a process known as cooperative affinity. The current research extends this observation to the use of heteromultimers. The composition of the multimers will be driven by DNA array and monomer ligand binding data, They will have advantages over agents directed against single epitopes by not relying on a single overexpressed cell surface protein. They will have advantages over other multifunctional agents in that they are produced via convergent synthesis and the multimer will be relatively small. Two hypotheses drive this work. 1) Heteromulfimeric ligands directed against combinations of cell surface proteins will bind with higher affinity and apparent cooperativity compared to individual monomers. 2) Combinations of expressed cell surface proteins can uniquely define individual cell types. The two aims of the proposed work follow these two hypotheses. Aim 1 will use a model system comprised of four well-defined ligand-receptor combinations. This system will be used to determine the optimum linker spacing and rigidity, the optimal avidity of the monomeric units, the in vitro binding curves and the in vivo biodistribution characteristic. Aim 2 will screen the expression patterns of 209 cell surface proteins associated with metastatic melanoma. This screen will use a custom DNA array, which will generate comprehensive data sets to identify potential targets. These data will be used to compare patterns of expression in melanoma cells from established and low passage primary cultures, to libraries ofmRNA from normal human tissues. These will be confirmed by RT-PCR, and by binding assays using bona fide agonists. At the end of Aim 2, the two aims will converge in the design of multimeric ligands that will specifically recognize metastatic melanoma for eventual use in detection, diagnosis and treatment of this terminal cancer.